- Video Analysis and Summarization
- Cutaneous Melanoma Detection and Management
- Sports Analytics and Performance
- Music and Audio Processing
- Medicine and Dermatology Studies History
- Advanced Image and Video Retrieval Techniques
- AI in cancer detection
- Image Retrieval and Classification Techniques
- Cutaneous lymphoproliferative disorders research
- Optical Coherence Tomography Applications
- Digital Imaging in Medicine
- Blockchain Technology Applications and Security
- Health Systems, Economic Evaluations, Quality of Life
- Nonmelanoma Skin Cancer Studies
- Social Media in Health Education
- Color perception and design
- Advanced Malware Detection Techniques
- Cloud Data Security Solutions
- Dermatologic Treatments and Research
- Chemotherapy-related skin toxicity
- Facial Rejuvenation and Surgery Techniques
- Media Influence and Health
- Spam and Phishing Detection
- Full-Duplex Wireless Communications
- Cancer Diagnosis and Treatment
Industrial University of Ho Chi Minh City
2024
The University of Texas MD Anderson Cancer Center
2022-2024
Mayo Clinic in Florida
2024
The University of Texas Health Science Center at Houston
2022-2023
Ho Chi Minh City University of Technology
2019
Vietnam National University Ho Chi Minh City
2019
IBM (United States)
2011-2018
IBM Research - Thomas J. Watson Research Center
2016-2018
Melanoma is the deadliest form of skin cancer. While curable with early detection, only highly trained specialists are capable accurately recognizing disease. As expertise in limited supply, automated systems identifying disease could save lives, reduce unnecessary biopsies, and costs. Toward this goal, we propose a system that combines recent developments deep learning established machine approaches, creating ensembles methods segmenting lesions, as well analyzing detected area surrounding...
The production of sports highlight packages summarizing a game's most exciting moments is an essential task for broadcast media. Yet, it requires labor-intensive video editing. We propose novel approach auto-curating highlights, and demonstrate to create first kind, real-world system the editorial aid golf tennis reels. Our method fuses information from players' reactions (action recognition such as high-fives fist pumps), expressions (aggressive, tense, smiling, neutral), spectators (crowd...
We present a system to assist users in dietary logging habits, which performs food recognition from pictures snapped on their phone two different scenarios. In the first scenario, called "Food context", we exploit GPS information of user determine restaurant they are having meal at, therefore restricting categories recognize set items menu. Such context allows us also report precise calories about meal, since chains tend standardize portions and provide each meal. second "Foods wild" try...
The production of sports highlight packages summarizing a game's most exciting moments is an essential task for broadcast media. Yet, it requires labor-intensive video editing. We propose novel approach auto-curating highlights, and use to create real-world system the editorial aid golf reels. Our method fuses information from players' reactions (action recognition such as high-fives fist pumps), spectators (crowd cheering), commentator (tone voice word analysis) determine interesting game....
We introduce a novel multi-modal system for auto-curating golf highlights that fuses information from players' reactions (celebration actions), spectators (crowd cheering), and commentator (tone of the voice word analysis) to determine most interesting moments game. The start highlight is determined with additional metadata (player's name hole number), allowing personalized content summarization retrieval. Our was demonstrated at Masters 2017, major tournament, generating real-time four live...
Melanoma is the deadliest form of skin cancer. While curable with early detection, only highly trained specialists are capable accurately recognizing disease. As expertise in limited supply, automated systems identifying disease could save lives, reduce unnecessary biopsies, and costs. Toward this goal, we propose a system that combines recent developments deep learning established machine approaches, creating ensembles methods segmenting lesions, as well analyzing detected area surrounding...
Rapid growth in the capture and generation of images videos is driving need for more efficient effective systems analyzing, searching, retrieving this data. Specific challenges include supporting automatic content indexing at a large scale accurately extracting sufficiently number relevant semantic concepts to enable search. In paper, we describe development system massive-scale visual concept extraction learning video. The models space using hierarchical faceted classification scheme across...
Electric load forecasting is crucial in a power system comprising electricity generation, transmission, distribution, and retail. Due to its high accuracy, the ensemble learning method XGBoost has been widely applied forecasting. XGBoost's performance depends on hyperparameters Genetic Algorithm (GA) commonly used algorithm determining optimal for this model. In study, we propose flowchart investigate impact of GA parameters accuracy models over hyperparameter grid The maximum data...
Access to social networks via portable devices promises be especially valuable for public communities. Because smart phones increasingly accompany community members wherever they are, the network extends into everyday interactions where location-specific, real time are natural and resulting actions can immediate. Governments NGOs leverage self-organizing, collectively-intelligent improve situational awareness community's resilience. At this of limited budgets, static but authoritative data...
Objectives: A discrepancy exists in the geographical distribution of dermatologists across United States (US). The aim this study was to reexamine current density describe any changes compared date from 5 and 12 years ago.
 Methods: Membership data 2021 American Academy Dermatology (AAD) database were retrieved characterize 3-digit zip code areas.
 Results: Out 712 populated codes with dermatologists, 510 (71.6%) had less than 4 per 100,000, as 515 (72.3%) 2016 (– 0.19% CAGR)....
Introduction: Recently, the Centers for Medicare and Medicaid services have emphasized increased reimbursement evaluation management with compensatory cuts in procedural services. A paucity of data exists examining within dermatology. Methods: This was a retrospective analysis databases maintained by Services which aimed at assessing if rates dermatology followed other specialties studied over time. The Physician/Supplier Procedure Summary File, total summary Part B carrier claims, queried...
Acroangiodermatitis (AAD)[KL1] is a rare vasoproliferative disorder often involving the extremities that has been classified into two variants. Mali-type AAD more common and associated with chronic venous stasis. Stewart-Bluefarb syndrome[KL2], other variant, underlying arteriovenous abnormalities. relatively benign diagnosis but it may mimic harmful etiologies such as Kaposi sarcoma both clinically histologically. A 67-year-old woman history of varicose veins, deep vein thrombosis, stroke,...
The production of sports highlight packages summarizing a game's most exciting moments is an essential task for broadcast media. Yet, it requires labor-intensive video editing. We propose novel approach auto-curating highlights, and use to create real-world system the editorial aid golf reels. Our method fuses information from players' reactions (action recognition such as high-fives fist pumps), spectators (crowd cheering), commentator (tone voice word analysis) determine interesting game....